-
Notifications
You must be signed in to change notification settings - Fork 226
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[SYCL] restructured cell linked list and other optimizations #623
Conversation
…or kernel executions
…ss for contiguous cells
To better comment on the new data structure for the cell linked list: |
Great! Thanks. I am current working on target of gradually merge your work to the main branch. For this, I am doing preparation changes so that I can implement SYCL for the entire library. If you have interest, you can check the branch sycl_redo. |
Is this change is aim a similar approach as Nvidia's implementation for particle method? |
Yes the approach is similar, but the main difference is that particles are directly positioned based on the cell size instead of sorting them based on their cell id
Thank you I will keep an eye out for it, you can ping me if there are questions on what I've implemented |
Sorry for the long PR, I have noticed some commits I didn't upstream before.
The main changes are:
Other changes are:
SPHINXSYS_USE_ONEDPL_SORTING
, which should be added to the CMake if this addition is deemed valuable, otherwise I could revert the commit and/or move it to a different PR for discussion